October 2010 - Pharo-dev - lists.pharo.org

[Pharo-project] Delay when I upload to Squeaksource 2MB files size

by nullPointer

Seems upload correctly, but when the progress bar is almost ended seems much more slow... Regards. -- View this message in context: http://forum.world.st/Delay-when-I-upload-to-Squeaksource-2MB-files-size-tp… Sent from the Pharo Smalltalk mailing list archive at Nabble.com.

Oct. 27, 2010

Re: [Pharo-project] posting etiquette - please post new text at the *top* of your email

by StÃ©phane Ducasse

+ 1 > Hi Folks, > > It is really hard keeping up with the deluge of email on this list. It is even harder if you have to search through posts to find the actual new text buried in all the quoted stuff. > > Unless there is a very good reason to provide the context, please add your new text *before* any quoted material. > > Thanks a bunch.

Oct. 27, 2010

Re: [Pharo-project] OCompletion doesn't work in Pharo 1.2 because of Preferences delete

by StÃ©phane Ducasse

thanks! On Oct 27, 2010, at 3:52 AM, Levente Uzonyi wrote: > On Tue, 26 Oct 2010, Mariano Martinez Peck wrote: > >> HI Ramain (or someone with time). >> >> OCompletion does not work anymore in Pharo 1.2 because of the Preferences >> removal. It would be nice if you could update it to use the new settings >> framework. > > Ocompletion-ul.74 should work in both Pharo 1.1 and 1.2. After loading the package evaluate [ECToolSet register]. > > > Levente > >> >> There is information here: >> http://book.pharo-project.org/book/CustomizingPharo/DeclaringSetting/ >> >> Thanks >> >> mariano >> > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

Oct. 27, 2010

[Pharo-project] posting etiquette - please post new text at the *top* of your email

by Oscar Nierstrasz

Hi Folks, It is really hard keeping up with the deluge of email on this list. It is even harder if you have to search through posts to find the actual new text buried in all the quoted stuff. Unless there is a very good reason to provide the context, please add your new text *before* any quoted material. Thanks a bunch. - on

Oct. 27, 2010

Re: [Pharo-project] akuhn/SUnit: run faster tests first & step through tests

by Adrian Kuhn

Thanks to Niko for campaigning me as maintainer :) I do feel very excited about unit testing in Pharo, however these days I gotta concentrate on my thesis only. Please feel free to reuse whatever is of use to you. My code is there to be cloned and forked :) --AA On Oct 26, 2010, at 12:01 , StÃ©phane Ducasse wrote: > On Oct 25, 2010, at 10:12 AM, Niko Schwarz wrote: > >> Couldn't Adrian be the maintainer? I could step in, too. > > yes but it means being present and reactive. > >> I don't really know what you have in mind for a testing framework. We >> have our own independent and fully compatible fork of SUnit: >> akuhn/SUnit, which is quite a strong code basis. > > but do we need should, class factory..... > >> And on top of that we >> have Phexample, which has its own >> name, some extra features and gives a clear point where Pharo is >> different from the rest. > > This is not the difference that we look for. This is addressing > people problems. > >> Just a suggestion: Let Adrian maintain akuhn/SUnit, perhaps minus the >> "should" syntax that you don't like (as in: #(a b c) size should = >> 3), >> and make akuhn/SUnit part of Pharo core. The new features (having >> tests depend >> on one another, "should" syntax) go into Phexample, and I could >> maintain that. > > Why not. If adrian wants. > Now first we should have > - a better name > - clean the code = remove class factory + add comments > - make sure that the tools can work with it > - move the long execution Sunit extension to it > - make sure that the tests are green > - integrate/digest latest Sunit feature > > >> >> Cheers, >> >> Niko >> >> On Sun, Oct 24, 2010 at 8:37 AM, StÃ©phane Ducasse >> <stephane.ducasse(a)inria.fr> wrote: >>> Lack of maintainers, too many unwanted features, lack of >>> explanation, >>> large changes... >>> >>> And this is not that several persons did not look at the code. It >>> got more than the average. >>> >>> Now I'm thinking that with discussion and care we should really >>> get another testing framework with the same API that SUnit >>> because if we want that PharoSunit is the same as Sunit then we >>> are doomed to be backward compatible and no >>> real improvements. >>> >>> Stef >>> >>> On Oct 23, 2010, at 10:58 PM, Adrian Kuhn wrote: >>> >>>> Sad to hear that. >>>> >>>> Technical difficulties, or just lack of customers? ^^ >>>> >>>> --AA >>>> >>>> On Oct 23, 2010, at 08:05 , StÃ©phane Ducasse wrote: >>>> >>>>> It was too difficult to integrate. >>>>> >>>>> Stef >>>>> >>>>> On Oct 23, 2010, at 2:01 PM, Niko Schwarz wrote: >>>>> >>>>>> What's the status of akuhn/SUnit? I couldn't find his code in >>>>>> 1.2 core. >>>>>> >>>>>> Niko >>>>>> >>>>>> On Wed, Feb 3, 2010 at 1:13 PM, StÃ©phane Ducasse >>>>>> <stephane.ducasse(a)inria.fr> wrote: >>>>>>> this loads well in the latest 1,1 core :) >>>>>>> so now we should really have a look and these cool changes >>>>>>> >>>>>>> Stef >>>>>>> >>>>>>> On Jan 6, 2010, at 2:38 PM, Adrian Kuhn wrote: >>>>>>> >>>>>>>> SystemChangeNotifier uniqueInstance >>>>>>>> noMoreNotificationsFor: TestCase. >>>>>>>> Gofer it >>>>>>>> disablePackageCache; >>>>>>>> squeaksource: 'akuhn'; >>>>>>>> package: 'SUnit'; >>>>>>>> package: 'SUnitGUI'; >>>>>>>> load. >>>>>>> >>>>>>> >>>>>>> _______________________________________________ >>>>>>> Pharo-project mailing list >>>>>>> Pharo-project(a)lists.gforge.inria.fr >>>>>>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >>>>>>> >>>>>> >>>>>> >>>>>> >>>>>> -- >>>>>> http://scg.unibe.ch/staff/Schwarz >>>>>> twitter.com/nes1983 >>>>>> Tel: +41 076 235 8683 >>>>>> >>>>>> _______________________________________________ >>>>>> Pharo-project mailing list >>>>>> Pharo-project(a)lists.gforge.inria.fr >>>>>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >>>>> >>>> >>> >>> >>> _______________________________________________ >>> Pharo-project mailing list >>> Pharo-project(a)lists.gforge.inria.fr >>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >>> >> >> >> >> -- >> http://scg.unibe.ch/staff/Schwarz >> twitter.com/nes1983 >> Tel: +41 076 235 8683 >> >> _______________________________________________ >> Pharo-project mailing list >> Pharo-project(a)lists.gforge.inria.fr >> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >

Oct. 27, 2010

Re: [Pharo-project] OCompletion doesn't work in Pharo 1.2 because of Preferences delete

by Levente Uzonyi

On Tue, 26 Oct 2010, Mariano Martinez Peck wrote: > HI Ramain (or someone with time). > > OCompletion does not work anymore in Pharo 1.2 because of the Preferences > removal. It would be nice if you could update it to use the new settings > framework. Ocompletion-ul.74 should work in both Pharo 1.1 and 1.2. After loading the package evaluate [ECToolSet register]. Levente > > There is information here: > http://book.pharo-project.org/book/CustomizingPharo/DeclaringSetting/ > > Thanks > > mariano >

Oct. 27, 2010

Re: [Pharo-project] Smells looking at collections

by Eliot Miranda

Hi Florin, On Tue, Oct 26, 2010 at 3:26 PM, Florin Mateoc <fmateoc(a)gmail.com> wrote: > On 10/26/2010 2:27 PM, Eliot Miranda wrote: > > > > The argument is not over the utility of stateful traits. The argument > is over how to access that state. The > > existing Traits implementation allows indirect access to state, where the > Trait uses accessors that must be > > implemented by a class using the trait. This has the disadvantages > listed by Levente, possibility of accessor > > selectors clashing with existing protocol, lack of privacy and poorer > performance. One possible solution is to do a > > new traits implementation that supports state, but as discussed > integration is hard and that could be a lot of work. > > There are other solutions to explore that could localise the changes. > > > > For example imagine adding some special illegal syntax that a variant of > the compiler (a compiler mode or a subclass > > or...) can deal with and using that to reference trait state. Then > changes are localised to > > - the compiler (to compile the syntax down onto the relevant inst vars) > > - class definition/trait composition (to ensure the state inst vars exist > in the class) > > - shout (to highlight the new syntax) > > - the decompiler (so decompiled trait code displays the new syntax for > trait state access) > > > > Another possibility might be a selector privacy hack. The selectors used > to access state in a trait are known not to > > clash in the trait (they ca be so chosen by the trait author not to > conflict with anything in the trait) but they > > cannot be guaranteed not to clash with a user of the trait. So > post-compile the selectors for the trait's state could > > be changed to private values (keys that existed only in the method > dictionary of the trait user). This doesn't have > > the performance of direct inst var access, but it solves the clash and > privacy issues. > > > > And I expect there are other solutions too. > > > > In summary don't argue for or against stateful traits, instead think > about how to add state to traits whilst having > > the most minimal impact on the rest of the infrastructure. > > > > best > > Eliot > > > > For me, the utility argument is an important one. I don't care about just > making it work, whether I believe in it or not. > It's bad enough that I have to do that at my workplace. Then again, they > are paying me for it. > Whatever. I think you're taking me wrong. I agree that stateful traits are more useful than non. I'm trying to discuss the engineering of them. Are you interestined in realising something you want or are you more interested in scoring points? Or is getting things to work not what you get paid for? Eliot > Cheers, > Florin > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >

Oct. 26, 2010

Re: [Pharo-project] Smells looking at collections

by Florin Mateoc

On 10/26/2010 2:27 PM, Eliot Miranda wrote: > > The argument is not over the utility of stateful traits. The argument is over how to access that state. The > existing Traits implementation allows indirect access to state, where the Trait uses accessors that must be > implemented by a class using the trait. This has the disadvantages listed by Levente, possibility of accessor > selectors clashing with existing protocol, lack of privacy and poorer performance. One possible solution is to do a > new traits implementation that supports state, but as discussed integration is hard and that could be a lot of work. > There are other solutions to explore that could localise the changes. > > For example imagine adding some special illegal syntax that a variant of the compiler (a compiler mode or a subclass > or...) can deal with and using that to reference trait state. Then changes are localised to > - the compiler (to compile the syntax down onto the relevant inst vars) > - class definition/trait composition (to ensure the state inst vars exist in the class) > - shout (to highlight the new syntax) > - the decompiler (so decompiled trait code displays the new syntax for trait state access) > > Another possibility might be a selector privacy hack. The selectors used to access state in a trait are known not to > clash in the trait (they ca be so chosen by the trait author not to conflict with anything in the trait) but they > cannot be guaranteed not to clash with a user of the trait. So post-compile the selectors for the trait's state could > be changed to private values (keys that existed only in the method dictionary of the trait user). This doesn't have > the performance of direct inst var access, but it solves the clash and privacy issues. > > And I expect there are other solutions too. > > In summary don't argue for or against stateful traits, instead think about how to add state to traits whilst having > the most minimal impact on the rest of the infrastructure. > > best > Eliot > For me, the utility argument is an important one. I don't care about just making it work, whether I believe in it or not. It's bad enough that I have to do that at my workplace. Then again, they are paying me for it. Cheers, Florin

Oct. 26, 2010

Re: [Pharo-project] Fwd: Cog VM doesn't start to listen on socket

by Tudor Girba

Ahh, I see :) Doru On 26 Oct 2010, at 22:14, Lukas Renggli wrote: > No, I fixed it in Kom. > > Lukas > > On 26 October 2010 22:12, Tudor Girba <tudor.girba(a)gmail.com> wrote: >> So, you always have to pass a script that restarts the server? This sounds like a workaround, right? >> >> Cheers, >> Doru >> >> >> On 26 Oct 2010, at 21:48, Lukas Renggli wrote: >> >>> Ok, I think I solved the issue. Restarting the server on startup seems >>> to solve the problem on all my machines. >>> >>> Thanks again for the hints! >>> >>> Lukas >>> >>> On 26 October 2010 20:41, Lukas Renggli <renggli(a)gmail.com> wrote: >>>> Thank you Eliot for the answer. I will have to investigate and see if >>>> this is a Seaside specific issue or related to Kom. >>>> >>>> Lukas >>>> >>>> 2010/10/26 Eliot Miranda <eliot.miranda(a)gmail.com>: >>>>> >>>>> >>>>> ---------- Forwarded message ---------- >>>>> From: Eliot Miranda <eliot.miranda(a)gmail.com> >>>>> Date: Tue, Oct 26, 2010 at 10:23 AM >>>>> Subject: Re: Cog VM doesn't start to listen on socket >>>>> To: Lukas Renggli <renggli(a)gmail.com> >>>>> >>>>> >>>>> Hi Lukas, >>>>> obviously we use sockets intensively in Teleplace and we see no problems >>>>> that are to do with Cog. But Cog does throw up timing issues in your code >>>>> because it has very different performance to the base VM (Smalltalk code is >>>>> faster, most primitives have the same speed, allocation is slightly faster, >>>>> etc). So my gut feeling is that what you're seeing is a timing related >>>>> issue (race condition?) that is revealed when on Cog and hidden when on the >>>>> normal VM. >>>>> HTH (yeah, right... ed.) >>>>> Eliot >>>>> >>>>> On Tue, Oct 26, 2010 at 5:04 AM, Lukas Renggli <renggli(a)gmail.com> wrote: >>>>>> >>>>>> Hi Eliot, >>>>>> >>>>>> I noticed that on the Cog VM the Kom Seaside server does not >>>>>> automatically start to listen, even if there is a process waiting at >>>>>> the semaphore of the socket. Singaling that semaphore from the process >>>>>> browser fixes the problem and the Cog image is fully working until the >>>>>> next time it is opened. >>>>>> >>>>>> The exact same image opened with an old VM does not show the problem >>>>>> and immediately starts to listen by itself. >>>>>> >>>>>> Did anybody else notice the same problem? Anybody found a solution? >>>>>> >>>>>> Lukas >>>>>> >>>>>> -- >>>>>> Lukas Renggli >>>>>> www.lukas-renggli.ch >>>>> >>>>> >>>>> >>>>> _______________________________________________ >>>>> Pharo-project mailing list >>>>> Pharo-project(a)lists.gforge.inria.fr >>>>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >>>>> >>>> >>>> >>>> >>>> -- >>>> Lukas Renggli >>>> www.lukas-renggli.ch >>>> >>> >>> >>> >>> -- >>> Lukas Renggli >>> www.lukas-renggli.ch >>> >>> _______________________________________________ >>> Pharo-project mailing list >>> Pharo-project(a)lists.gforge.inria.fr >>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >> >> -- >> www.tudorgirba.com >> >> "Sometimes the best solution is not the best solution." >> >> >> _______________________________________________ >> Pharo-project mailing list >> Pharo-project(a)lists.gforge.inria.fr >> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >> > > > > -- > Lukas Renggli > www.lukas-renggli.ch > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project -- www.tudorgirba.com "Problem solving efficiency grows with the abstractness level of problem understanding."

Oct. 26, 2010

Re: [Pharo-project] OCompletion doesn't work in Pharo 1.2 because of Preferences delete

by StÃ©phane Ducasse

On Oct 26, 2010, at 11:38 PM, Levente Uzonyi wrote: > On Tue, 26 Oct 2010, StÃ©phane Ducasse wrote: > >> >> On Oct 26, 2010, at 11:18 PM, Levente Uzonyi wrote: >> >>> On Tue, 26 Oct 2010, Mariano Martinez Peck wrote: >>> >>>> HI Ramain (or someone with time). >>>> >>>> OCompletion does not work anymore in Pharo 1.2 because of the Preferences >>>> removal. It would be nice if you could update it to use the new settings >>>> framework. >>> >>> I fixed some of the preference issues (i didn't upload it yet), but it's a PITA to make it work in Pharo 1.2, because too many things were changed and are therefore broken. >> >> I can imagine. >> We should really get a way to support automatic change application but this is a real job to build that. >> >>> Btw what's with the "keep deprecated stuff for the next 2 minor versions" policy? >> >> this is what we do: we will remove 1.1 deprecated methods in 1.3 alpha. May be we did not apply it totally and are messy with that >> because we did not deprecated all the preferences methods. >> >>> Preferences are gone in 1.2, but Settings were only introduced in 1.1 and Preferences doesn't seem to be deprecated at all in 1.1. So, according to the policy, Preferences should be available till at least 1.3. >> >> the preferences class is just there in 1.2 to help people loading their code. but the behavior is gone. > > No, it's not there. At least not in the core image (12216). then this is just a mistake. > > > Levente > >> >> Settings were introduced in 1.1 (but not used) so that people can migrate their code in 1.1 to save it in 1.2. >> Alain reminds that recently. >>> >>> >>> Levente >>> >>>> >>>> There is information here: >>>> http://book.pharo-project.org/book/CustomizingPharo/DeclaringSetting/ >>>> >>>> Thanks >>>> >>>> mariano >>>> >>> >>> _______________________________________________ >>> Pharo-project mailing list >>> Pharo-project(a)lists.gforge.inria.fr >>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >> >> >> _______________________________________________ >> Pharo-project mailing list >> Pharo-project(a)lists.gforge.inria.fr >> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

Oct. 26, 2010